US5949355A - Method and apparatus for adaptive data compression - Google Patents
Method and apparatus for adaptive data compression Download PDFInfo
- Publication number
- US5949355A US5949355A US08/915,961 US91596197A US5949355A US 5949355 A US5949355 A US 5949355A US 91596197 A US91596197 A US 91596197A US 5949355 A US5949355 A US 5949355A
- Authority
- US
- United States
- Prior art keywords
- data
- record
- bit
- data elements
- repeat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
- H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
Definitions
- the present invention relates generally to data storage and communication, and particularly to systems for compressing data to facilitate denser data storage and faster data communication.
- the present invention consists of a number of specialized data compression subsystems--each designed to compress a particular type of data--that are utilized to implement an adaptive, data-indifferent compression system that has achieved an average compression ratio for all files and data types in excess of 3:1.
- the compression system made possible through the present invention adapts to any type of data by combining to achieve optimum overall compression. Tested over a range of files including bitmap, database, spreadsheet, ASCII, EBCDIC, and text/word processor files, this data compression system has achieved compression ratios ranging from 2.7:1 to nearly 10:1.
- the data compression systems of the present invention can be used to compress data in any language, text or binary format, regardless of its machine language encoding.
- the data compression system of the present invention is implemented in three main functional groups: the scanners, the decision engine, and the compression modules.
- the scanners are implemented in two stages, called Copy Scan and Main Scan.
- the scanners handle input from a data file.
- the file can be either a data storage file or a data communication file.
- the file consists of a stream of bytes or characters, a byte typically containing eight (8) bits of data.
- the file is divided into records (in the preferred embodiment, a record is 2048 bytes long) by the user or application, and the scanner then processes the file one record at a time.
- the scanners examine the data and produce a data profile report.
- the data profile report is a detailed breakdown of the different types of data characters and groups in the record.
- the report indicates the identity of each byte of data and classifies each byte into one of eight categories.
- the report gives the frequency with which specific characters and groups of characters occur in the record and the relative position of those characters and groups.
- the report also indicates the identity and location of characters, words and phrases that are repeated within the record.
- the decision engine processes the data profile report and classifies the entire record as fitting into one of eight data composition profiles, or modes.
- the decision acts according to a set of statistical rules.
- the decision engine constructs a program for optimum compression of the record.
- the program made by the decision engine consists of detailed instructions regarding application of the compression modules.
- Each compression module is a highly specialized compression subsystem designed to compress a specific type of data (E.g., ASCII capitals, Hexadecimal numerics, etc.).
- the decision engine's program specifies which compression modules are to be applied and the order in which they are to be applied, as well as the conditions that will mandate branching from one module to another for better compression.
- the modules then compress the data, branching among themselves according to the instructions provided by the decision engine.
- execution of the compression modules is dynamically adaptive (following the "road map" provided by the decision engine) for optimum compression.
- FIG. 1 shows the table VALTBL which is used in text compression.
- FIG. 2 shows the table SCNTBL which is used in text compression.
- FIG. 3 shows the data structures Event Matrix and Event List which are used for block coding.
- Data compression through the system of the present invention involves five major steps: (1) Copy Scan; (2) Initial Evaluation; (3) Main Scan; (4) Statistics/Decision Engine; and (5) Compression. Each step is described below.
- the data file is opened and one record is read into a source buffer by the user or application.
- the application also allocates a 16 kilobyte work area for use by the compression system and prepares a parameter field in the first 16 bytes of the work area.
- the parameter field is the device through which the application and the compression system communicate with each other regarding the address of the source buffer, the length of the uncompressed data record (usually 2048 bytes), the size of the compressed record, and the location of the output.
- the input stage of the compression system performs the initial scan, called the Copy Scan.
- the Copy Scan is a data copy, identification and report loop.
- the Copy Scan uses the 16 kilobyte work area allocated by the user or application.
- the work area is divided into sections for variables, report fields, vectors, buffers and mapping sections.
- the work area sections used during the Copy Scan are the DATA COPY buffer, the QUE/ID buffer, and the Group Report Table.
- the compression system copies the record to a buffer and scans it to identify and characterize its contents.
- the record is copied sequentially to CPYBUF (the DATA COPY buffer).
- CPYBUF the DATA COPY buffer
- a counter in the Group Report Table corresponding to the particular character detected is incremented, thereby tracking the number of times each individual character occurs in the record (i.e., the number of As, the number of spaces, the number of bs, etc.).
- Each byte is also classified according to which of eight data groups it belongs, as determined by the character groupings defined in the internal table IDTBL (identification table), as set forth below in Table 1.
- the IDTBL value corresponding to the class identification for each byte is stored sequentially in QUEBUF, 2048 bytes from the corresponding character identification code in CPYBUF (these values will be replaced in a later scan by codes calling specific compression modules and indicating the number of characters in a word or number).
- a counter corresponding to that category is incremented in the Group Report Table.
- the eighth bit state of each byte is tracked in order to detect if the record is comprised of full range binary data.
- the system uses a logical OR operation on the eighth bit to detect a HIGH value in the eighth bit; if one is detected, then other checks are executed to determine if the data is full range binary.
- the ASCII group counters are checked for a low or zero aggregate count, and the EBCDIC group counters are checked for a high aggregate count. If these conditions are found, then the data is identified as EBCDIC and translated to ASCII equivalents to simplify compression.
- a flag is set to indicate the translation from EBCDIC, so that the decompression system will know to translate the data back to EBCDIC.
- the identification of the record as ASCII or full range binary is also used to set the Range Limiters used during the Main Scan. For ASCII, the Range Limiters are set at 128; for full range binary they are set at 256.
- Copy Scan activates a check to detect contiguous repetition of characters and character pairs. If Copy Scan detects a byte as being a repetition of the preceding byte, then it branches to the Repeat Scan subroutine. The Repeat Scan subroutine then scans the entire record to detect repeat events. The Repeat Scan subroutine then stores codes in the Que Buffer and Auxiliary Buffer indicating the location and length of each repetition found, and also updates the corresponding Repeat Event Counters in the Group Report Table. The Repeat Scan subroutine does not update the counts of individual characters or character groups, nor does it continue to execute the eighth-bit OR operation. When the Repeat Scan subroutine is finished, it branches back to the Copy Scan. Copy Scan then finishes scanning the non-repetitive bytes in the record.
- the Initial Evaluation step uses the information gathered during Copy Scan to make adjustments to the Main Scan. These adjustments ensure that the Main Scan proceeds more quickly and efficiently.
- the identification steps executed in Copy Scan provide useful information regarding the makeup of the record, such as the existence of ASCII and EBCDIC groups, which classes of characters (per IDTBL) are used and the individual characters used in each class. Based on this information, the Initial Evaluation can designate what types of data the Main Scanner should look for and what types of data it should ignore, or drop. The designation of what to look for and what to drop is in terms of the eight data classifications in IDTBL Thus, for example, if the counters in the Group Report Table corresponding to ASCII capital letters (character ID 04h) are low, the Initial Evaluation might designate that ASCII capital letters be dropped and not looked for during the Main Scan. In that event, any ACSII capital letters contained in the record will be encoded and compressed as parts of block repetitions or some other repetition event, as special characters, or they might not be compressed at all.
- the Main Scan completes the scanning of the record that was commenced by Copy Scan.
- the Main Scan has fewer groups to identify, so it is quicker and more complete than the Copy Scan.
- the Main Scan updates the character frequency table (which was not updated during the Copy Scan to account for characters present in the repetition events scanned by the Repeat Scan subroutine), and other counters in the Group Report Table that were not updated during the Copy Scan.
- the Main Scan also executes the character blocking function. In the character blocking function, the Main Scan detects multiple occurrences (as contrasted with contiguous repetitions) of the same block of characters (e.g., the word "the”) and encodes them as block repetitions.
- the information gathered during Main Scan is compiled into a detailed report. From the report, the Decision Engine will know the record's contents in detail--in terms of capital words, lower case words, mixed words, numbers, hexadecimal numbers, repetition events, and symbols. The report also gives a count of the number of times individual characters are found in the record (E.g., 128 spaces, 25 A's, etc.).
- the information in the report compiled by the Main Scan is used by the next system stage, namely the Statistics/Decision Engine stage, to determine the next step; i.e., to evaluate which compression subsystems are likely to achieve the best compression.
- the Statistics/Decision Engine step contains the intelligence of the compression system. Based on the information compiled during the preceding scans, the Decision Engine establishes the priority of application of the individual compression modules.
- the input to the Decision Engine is the report of the record's contents prepared by the scanning operations. Based on the record's contents, the Decision Engine assigns one of eight modes to the record, as indicated in Table 2:
- Mode 00 is the default mode, where the major record contents (not counting repetition and block events) are capital letters and the minor record contents are symbols.
- Mode 01 is assigned where the major record contents are capital letters and the minor record contents are lower case letters and symbols.
- Mode 02 is assigned where the major record contents are lower case letters, and the minor record contents are symbols.
- Mode 03 is assigned where the major record contents are lower case letters and the minor contents are capital letters and symbols.
- Mode 04 is assigned where the major record contents are either non-hexadecimal numbers or hexadecimal numbers and the minor record contents are symbols.
- Mode 05 is assigned where the major record contents are non-hexadecimal numbers and capital letters and the minor contents are symbols.
- Mode 06 is assigned where the major record contents are non-hexadecimal numbers and lower case letters and the minor contents are symbols.
- Mode 07 is assigned where there is an equal or random mix of capital and lower case letters, numbers and symbols.
- the Decision Engine also selects special characters (e.g., if a certain character occurs frequently in the record, a special four-bit compression code can be assigned to that character instead of its eight-bit ASCII code), builds a header that is inserted at the beginning of the compressed record (the header describes the types of compression applied, for use by the decoder), refines the block coding done in the Main Scan, and translates EBCDIC to ASCII setting a translation flag as necessary.
- special characters e.g., if a certain character occurs frequently in the record, a special four-bit compression code can be assigned to that character instead of its eight-bit ASCII code
- the record's header indicates the record's mode and its compressed length.
- the header also contains sixteen prefix usage descriptor flags that indicate which character zones are present in the compressed record. Any zones that are not present are available for use as special character codes.
- the first bit in the mode field serves as a flag that indicates whether the data to follow is ASCII or not.
- the four-bit mode field also contains a one-bit flag that indicates whether the data is normal (ASCII) or reversed (translated) (BIN and EBCDIC). As an illustrative example, a zero (0) may be used to indicate ASCII and a one (1) would then indicate binary (BIN).
- the remaining three bits of the mode field indicate the particular mode assigned to the record by the decision engine for the purposes of compression. Since there are three bits that identify the mode, it should be clear that one of eight modes is possible. The assigned mode is based upon the statistical content of the record as analyzed by the scanner.
- the header descriptor field describes the assignment of coding prefixes and special character:
- Each of the sixteen flags in this field represents one zone or prefix
- zone 4 if a zone 4 value (i.e., 04 through F4) is coded as such in the compressed record, then the header descriptor flag corresponding to zone 4 (bit number 4) will indicate that zone 4 values are present in the compressed record, and accordingly, the value 4 cannot be used as a special character code. Conversely, if no zone 4 values are present in the record, then the value 4 may be assigned as a special character code.
- Zones 8-F are typically reserved as codes indicating various events that might be present in the record, such as character repetition, character pair repetition, and block repetition events (if no events corresponding to a particular zone occur in the record, then that zone is available for use as a different compression code, such as special character).
- the event codes are used as prefixes to indicate the type of coding that follows the prefix.
- the default values assigned to the event codes are as follows. Groupings of letters (words) and numbers are indicated by prefix values of 8, 9, and C. A prefix value of A is used to denote contiguous character repetition and pair repetition events. Non-contiguous repetition of a block of characters is indicated by prefix values of E and F.
- the remaining prefix values, B and D may be available a special character codes or additional values for indicating block repetition events.
- the descriptor field is used to assign special character code values.
- Special character codes are used to optimally compress certain characters, typically characters that appear frequently in the record. Using these codes, an eight-bit ASCII code for the character is compressed into a 4-bit special character code.
- a one (1) in a bit corresponding to one of the zones means that characters corresponding to the zone are coded in the compressed record, and therefore the zone is used in the compressed record and that value is not available for assignment as a special character code.
- a zero (0) means that the compressed record does not contain characters in the corresponding zone, nor (in the case of zones 8-F) does it contain an event whose prefix is in that zone, so the value corresponding to the bit containing the zero is available and may be assigned as a special character code.
- a list of the characters in the record that will be represented by special character codes is inserted following the descriptor field--one character per zero bit in the descriptor field.
- the value corresponding to the first zero bit in the descriptor field is assigned as the special character code for the first (sixteen bit ASCII) character in the special character list, the value of the second zero bit is assigned as the code for the second character in the list, and so on.
- the number sequence in the Header Descriptor indicates that zones 0, 2, 3, 4, 5, 6, 7, A, B, C, D, E, F are used as either character codes or event codes (prefixes) in the compressed record, and are not available to be assigned as special character codes.
- the values of the zero bits (1, 8 and 9), on the other hand, are available to be assigned as special character codes. This means that the compressed record does not contain coded character values in zone 1, nor does it includes any events whose prefixes are 8 or 9 (two of the default values for groupings of numbers and words).
- the ASCII codes for the characters that have been selected to be represented by special character codes are the ASCII codes for the characters that have been selected to be represented by special character codes.
- the first character is 20h (space), and will be compressed as 1 (0001); the second character, 2Ch (comma) will be compressed as 8 (1000); and the third character in the list, 2E (period or decimal point) will be compressed as 9 (1001).
- the sequence ",” is coded during compression as 81 (1000 0001), and the sequence ".” is coded as 911 (1001 0001 0001).
- the header is identical to the ASCII header described above, except that (1) a sixteen-bit extended header is added between the special character list and the start of the compressed data, and (2) a bit stream of undetermined length is added after the compressed data.
- the extended header is comprised of two four-bit binary mode flags, and a one-byte field to indicate bit stream length.
- the sixteen-bit extended header field which appears only in BIN (binary) mode, indicates whether or not special random process modules, and the prefixes assigned to them, are used. If the binary mode flags are set, then before applying any compression modules, the BIN record is converted to ASCII.
- the Decision Engine makes its decisions regarding application of the individual Compression Modules and transmits its instructions to the Compression Modules, the Decision Engine then calls the Compression Modules.
- the Decision Engine monitors the compression and, if a branch condition occurs, the Decision Engine directs the current module to branch to the next module, as determined by the priorities set by the Decision Engine.
- the Compression Modules execute compression of the record pursuant to the Decision Engine's instructions, branching among themselves as directed by the Decision Engine to maximize compression of the record.
- the compression modules may be separated into the following three groups for the purpose of description and discussion: Numerics, Text, and Repetition Event. Block repetitions are a subset of the Repetition Events.
- the numerics compression modules are designed for records that contain some numeric digits.
- the numerics compression modules are grouped into the major numerics modules, minor numerics modules, and hexadecimal numerics modules.
- the major numerics modules are used for records that are mostly non-hexadecimal numeric, i.e., modes 04, 05, and 06.
- the major numerics modules are NMBR08 and NMBR0A. These modules are designed to compress a number that contains trailing zeroes.
- the minor numerics modules are used for records that are mostly non-numeric but contain some non-hexadecimal numbers, i.e., mode 00, 01, 02, 03 and 07.
- the minor numerics modules are NMBR00, NMBR02, NMBR04, and NMBR06.
- the hexadecimal module is used for records that are mostly alpha-numeric digits, 0-9 and A-F, i.e., mode 04 records.
- NMBR00 This module is invoked for compressing non-hexadecimal numerics whose last digit is LOW--less than six (6).
- the default prefix used to indicate this compression module is 08h.
- the number is compressed by converting each digit (except the last one in the string) from its eight-bit ASCII code to its four-bit binary representation. For example, the number 4, whose ASCII code is 34h (0011 0100), would be coded as 4h (0100), which is four bits instead of eight.
- the last digit in the string is encoded by adding 10d to it; thus, a last digit of one (1) is encoded as Ah (1010).
- NMBR02. This module is invoked for compressing non-hexadecimal numerics whose last digit is HIGH--greater than five (5).
- the default prefix used to indicate this compression module is 09h.
- the number is compressed in the same manner as with NMBR00, except that a different value (4, 5 or 6) is added to the last digit in the string to ensure that the compressed string ends with a hexadecimal value in the range A through F.
- NMBR04. This module is invoked for compressing binary coded numerics whose value is less than 100.
- the default prefix used to indicate this compression module is 08h.
- NMBR06 This module is invoked for compressing binary coded numerics whose value is greater than 99.
- the default prefix used to indicate this compression module is 09h.
- NMBR0A This module is invoked for compressing long numerics followed by trailing zeroes.
- the compression is the same as for NMBR08.
- the default prefix used to indicate this compression module is 0Ch.
- LTRS00 and LTRS02. These modules are invoked for compressing pairs of letters, and are used where the record contains only or mostly capital letters (for LTRS00), or only or mostly lower case letters (for LTRS02).
- the default prefix used to indicate both of these compression modules is 08h.
- the modules are activated by the setting of a flag in an internal grouping of prefix values that are distinct from the prefixes that are used in coding.
- the internal prefix grouping is simply a set of internal flags used by the system to indicate which of several different coding schemes a given coding prefix will indicate.
- PRFX08 in the internal prefix group
- PRFX08 in the internal prefix group is set to a certain value which indicates that the coding prefix 08h is assigned to LTRS00 and LTRS02 coding for pairs of letters.
- the letters pair is encoded as PPPP CCCC CCCC, where P is the prefix (1000) and C is the ASCII code of the character in the pair.
- LTRS04 and LTRS06. These modules are invoked for compressing a letters triad (e.g., "AAA").
- the default prefix used to indicate these compression modules is 0Ch.
- LTRS00 and LTRS02 these modules are activated through the setting of a flag in the internal prefix grouping.
- PRFX0C is set to a certain value, the triad modules, LTRS04 and LTRS06 are activated.
- a letters triad is encoded as the prefix 0Ch followed by the ASCII code of the character in the triad.
- LTRS08 and LTRS0A These modules are invoked for compressing a mixed record consisting of all or mostly capital letters (LTRS08), or all or mostly lower case letters (LTRS0A).
- the default prefix used to indicate these compression modules is 09h.
- the flag-setting system described above in connection with LTRS00-LTRS06 is also used to activate these modules.
- the flag is set in PRFX09 of the internal prefix grouping.
- the text is compressed in the form PPPP FLLL CCCC . . . CCCC, where P is the prefix of 9h (1001), F is a flag indicating capitals or lower case, and C is the coded text.
- LTRS0C This module is invoked for compressing a fixed letter sequence that comprises a valid word.
- the default prefix used to indicate this compression module is 09h. The compression proceeds in the same manner as described above for LTRS08 and LTRS0A.
- the mechanisms for this compression are the tables VALTBL and SCNDTBL These are illustrated in FIGS. 1 and 2.
- the first letter of a word is compressed using VALTBL, and each subsequent letter in that word is compressed using SCNTBL.
- VALTBL is a list of 26 values representing each letter of the alphabet. Of the 26 values, sixteen are non-zero values and the remaining ten are zeroes. The zero values signify that the letters corresponding to them are not valid word-starting letters for that record, i.e., there are no or very few words in the record that start with these letters. Thus, the position of the zero values in VALTBL depends on the contents of the particular record being compressed.
- SCNTBL SCNTBL contains 26 fields, each field representing a letter of the alphabet, in sequence.
- Each field in SCNTBL contains 26 entries, each entry corresponding to a letter of the alphabet, in sequence.
- Each entry in a given field contains a value corresponding to the compression code for the next letter in the text string.
- VALTBL not all 26 letters of the alphabet are valid next letters in the text contained in the record being compressed. For example, for a given record the letter "B" may never be followed by the letter "X". In that event, the entry in the "B" field corresponding to "X" (i.e., the twenty-fourth entry) will contain a zero value.
- This scheme is based on the high probability that for a given record, no letter in the alphabet will have more than sixteen valid next letters following it. In this way, all next letters can be compressed as a four bit value between zero and fifteen.
- This module may be invoked, for example, for compressing the special case of a comma followed by a space. This combination (or some other frequently occurring combination) is compressed as FFh (1111 1111).
- Single Character Repeat This is a group of three modules designed to compress a text string consisting of a single character repeated three times or more.
- the three modules are designed for short, medium and long character repetition strings.
- the short character repeat module is invoked for 2-11 repetitions of the preceding character.
- the default prefix used to indicate this compression module i.e., "repeat the preceding character” is Ah (1010).
- the medium character repeat module is invoked for 9-12 repetitions of the preceding character.
- the default prefix used to indicate this compression module is 09h.
- repetition of a character 8 times is encoded as 8Fh (1000 1111).
- the ZNZB coding is PPPP XXXX XXXXXXX, where P is the prefix and the three "X" nibbles represent the three non-zero nibbles found in the uncompressed sequence.
- Block Repetition With reference to FIG. 3, block repetition encoding is accomplished through the use of three internal data structures, the Event Matrix, the Event List, and the Auxiliary Que Buffer.
- the Event Matrix contains 256 two-byte fields, corresponding to 00 through FF. Thus, each field in the Event Matrix represents a character that may possibly be a starting character in a block repetition event.
- the Event List contains a number of two-byte fields, each one corresponding to each character address (up to 2048) in the record being scanned.
- the Auxiliary Que Buffer has the same structure as the Event List.
- the corresponding Matrix address is updated to reflect the address of this most recent occurrence, and the corresponding Event list address will store the address of the previous occurrence.
- the value F8 would be stored at 0C in the Matrix, while address F8 in the Event List would store the value 0A (which had previously been in the corresponding Matrix field) in order to point to the previous occurrence of the block start. In this way, the block system points back to the character block that is to be repeated.
- the length of the block is stored in the Auxiliary Que Buffer at an addresses corresponding to each occurrence of the block event.
Abstract
Description
TABLE 1 ______________________________________ the IDTBL value Low Range Symbol 00d (00h) ASCII Numeric Digit 02d (02h) corresponding to ASCII Capital Letter 04d (04h) ASCII Lower Case Letter 06d (06h) the class High Range Symbol 08d (08h) EBCDIC Numeric Digit 10d (0Ah) identification for EBCDIC Capital Letter 12d (0Ch) EBCDIC Lower Case Letter 14d (0Eh) each byte is stored ______________________________________
TABLE 2 ______________________________________ MAJOR RECORD MINOR MODE CONTENTS RECORD CONTENTS ______________________________________ 00 CAPS SYM (Default) 01 CAPS LOWER andSYM 02LOWER SYM 03 LOWER CAPS/SYM 04 NUM orHEX SYM 05 NUM andCAPS SYM 06 NUM andLOWER SYM 07 CAPS/LOWER/NUM/SYM (Random in Binary) ______________________________________
______________________________________ Header Descriptor Characters To Be Assigned Sp. Ch. Codes ______________________________________ 1011 1111 0011 1111 0010 0000 0010 1100 0010 1110 ______________________________________
__________________________________________________________________________ Module and Description Prefix __________________________________________________________________________ NMBR00: Non-hex numeric digits with 08 To encode, add 10d to the last last digit LOW (<6). digit in the string. This ensures that the last coded digit is represent by A - F. When a value A - F is encountered, this indicates the end of the encoded non-hex numeric. NMBR02: Non-hex numeric digits with 09 To encode, add 4d (or 5d or 6d) to last digit HIGH (>5). the last digit in the string. NMBR04: Binary coded numeric <100. 08 NMBR06: Binary coded numeric >99. 09 NMBR08: Short numeric: with trailing 0C zeroes (e 10.sup.X). NMBR0A: Long numeric with trailing 0C zeroes (e 10.sup.X). LTRS00 (Caps Pair) and 08 LTRS02 (Lower Case Pair). LTRS04 (Caps Triad plus space) and 0C LTRS06 (Lower Case Triad). LTRS08 (Mixed Record Caps) and 09 or 0C LTRS0A (Mixed Record Lower plus space). LTRS0C (Fixed Letter Sequence). 09 LTRS0E (Special Case). 0C Character Repeat - Special 00-07 (0000 0ddd, where d = "don't care") Character Repeat - Not Special Short: 3-8 repetitions 8A-8F: 1000 10d + L (L = reps - 3) Example: Reps = 8 So: L = 8 - 3 = 5 Lo 4 bits: 10d + L = 10d + 5 = 15d (Fh) Code: 1000 1111, or 8F Medium: 9-12 repetitions 9A-9D: 1001 7d + L (L = reps - 6) Long: 13-44 repetitions 9E0-9FF: 1001 111L LLLL (L= reps - 13) Normal Repeat - Any Character Short: 2-11 repetitions A: 1010 LLLL (L = reps - 2) Med: 12-19 repetitions AA: 1010 1010 0LLL (L = reps - 10) Long: 20-147 repetitions AA8-AAF: 1010 1010 1LLL LLLL (L = reps - 8) Very Long: 148-2047 repetitions AB8-ABF: 1010 1011 1LLL LLLL LLLL(L = reps - 128) Pair Repeat - Special Pair 0F Pair Repeat - Normal Short: 1-4 repetitions AC-AF: 1010 12d + L (L = reps - 1) Medium: 5-68 repetitions AB80-ABBF: 1010 1011 10LL LLLL(L = reps - 5) Long: 69-1092 repetitions ABC00-ABFFF: 1010 1011 11LL LLLL LLLL (L = reps - 69) Pair Repeat - Relative: Two or more PPPP FSDD DDDD FSDD DDDD . . . matching character pairs within 63 FSDD DDDD bytes of one another. P = prefix F = flag S = sign D = difference Quad Repeat - Relative: Two or more PPPP FSDD DDDD DDDD FSDD matching groups of four characters DDDD . . . FSDD DDDD DDDD within 1023 bytes of one another. Low Relative Repeat: Groups of five PPPP F SDD SDD SDD SDD . . . F matching characters, each SDD SDD SDD SDD . . . F character being within 3 bytes of the previous occurrence. May be longer, but always a multiple of five. ZNZB (Zero/Nibble/Zero/Byte). PPPP XXXX XXXX XXXX Sequence of four bytes in the format:00h 0Xh 00h XXh. Eliminate the zero nibbles. Special Character Zero (0) in one of the sixteen bits in the header descriptor field; bit no. is the special character code for the corresponding character in the list following the header descriptor field (E.g., if bit no. 8 = 0, and it is the first zero in the field, then the special character code for the first character listed after the field is 8 -- if the first character listed is a space, then 8 = space). Block-to-Block Repeat (no displacement) Short (BL < 9) B0-B7: 1011 0LLL (L = BL - 1) Long (8 < BL < 138) B80-BFF: 1011 1LLL LLLL (L = BL - 9) Normal Block Repeat Very Short Block / Very Short D00-DFF: 1101 LLLD DDDD Displacement L = BL - 3 (2 < BL < 11 / 0 < BD < 33) D = BD - 1 Short Block / Short Displacement E000-EFFF: 1110 LLLL DDDD (BL < 19 / BD < 256) DDDD L = BL - 3 D = BD - 1 Short Block / Long Displacement F0000-FFFFF: (BL <19 / BD < 2049) 1111 0LLL LDDD DDDD DDDD L = BL - 3 D = BD Medium Block/ Short F8000-F9FFF: Displacement 1111 100L LLLL DDDD DDDD (18 < BL < 60 / BD < 256) L = BL - 19 D = BD Medium Block / Med. FA LLLL LDDD DDDD Displacement L = BL - 19 (18 < BL < 60 / 255 < BD < D = BD - 256 384) Medium Block / Long FB LLLL LDDD DDDD DDDD Displacement L = BL - 19 D = BD (18 < BL < 60 / 383 < BD < 2049) Med. Long Block / Any Disp. FC . . . - FD . . . (31 < BL < 96) 1111 110L LLLL LDDD DDDD DDDD L = BL - 32 Long Block / Any Displacement FE 0LLL LLLL LDDD DDDD DDDD (63 < BL < 319) L = BL - 64 Very Long Block / Any Disp. FE 10LL LLLL LLLL LDDD DDDD (318 < BL < 2049) DDDD L = BL - 319 __________________________________________________________________________
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US08/915,961 US5949355A (en) | 1994-12-06 | 1997-08-21 | Method and apparatus for adaptive data compression |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US08/350,426 US5684478A (en) | 1994-12-06 | 1994-12-06 | Method and apparatus for adaptive data compression |
US08/915,961 US5949355A (en) | 1994-12-06 | 1997-08-21 | Method and apparatus for adaptive data compression |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/350,426 Continuation US5684478A (en) | 1994-12-06 | 1994-12-06 | Method and apparatus for adaptive data compression |
Publications (1)
Publication Number | Publication Date |
---|---|
US5949355A true US5949355A (en) | 1999-09-07 |
Family
ID=23376670
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/350,426 Expired - Fee Related US5684478A (en) | 1994-12-06 | 1994-12-06 | Method and apparatus for adaptive data compression |
US08/915,961 Expired - Fee Related US5949355A (en) | 1994-12-06 | 1997-08-21 | Method and apparatus for adaptive data compression |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US08/350,426 Expired - Fee Related US5684478A (en) | 1994-12-06 | 1994-12-06 | Method and apparatus for adaptive data compression |
Country Status (1)
Country | Link |
---|---|
US (2) | US5684478A (en) |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6380873B1 (en) * | 2000-06-30 | 2002-04-30 | Quantum Corporation | Method for reducing radio frequency emissions on high-speed serial buses |
US20030110296A1 (en) * | 2001-12-07 | 2003-06-12 | Kirsch Steven T. | Method and system for reducing network latency in data communication |
US20030164020A1 (en) * | 2000-07-29 | 2003-09-04 | Haberkamm Klaus Dieter | Method and device for band-edge orientated displacement of intermediate cylinders in a 6 cylinder frame |
US20030174732A1 (en) * | 1999-03-01 | 2003-09-18 | Hughes Electronics Corporation | Technique for data compression by decoding binary encoded data |
US20040006455A1 (en) * | 2002-07-03 | 2004-01-08 | Vadim Fux | System and method of creating and using compact linguistic data |
US6717580B1 (en) * | 2000-02-02 | 2004-04-06 | America Online, Inc. | Method and apparatus using a list of <minimum, size, gap, offset > quadruple to encode unicode characters in an upper/lowercase mapping |
US20060090033A1 (en) * | 2004-10-22 | 2006-04-27 | International Business Machines Corporation | Facilitating Server Response Optimization |
US20060181441A1 (en) * | 1998-12-11 | 2006-08-17 | Fallon James J | Content independent data compression method and system |
US20060190644A1 (en) * | 2000-02-03 | 2006-08-24 | Fallon James J | Data storewidth accelerator |
US7148824B1 (en) * | 2005-08-05 | 2006-12-12 | Xerox Corporation | Automatic detection of character encoding format using statistical analysis of the text strings |
US7420992B1 (en) | 2005-03-17 | 2008-09-02 | Packeteer, Inc. | Adaptive network traffic compression mechanism including dynamic selection of compression algorithms |
US7649909B1 (en) | 2006-06-30 | 2010-01-19 | Packeteer, Inc. | Adaptive tunnel transport protocol |
US7777651B2 (en) | 2000-10-03 | 2010-08-17 | Realtime Data Llc | System and method for data feed acceleration and encryption |
US8054879B2 (en) | 2001-02-13 | 2011-11-08 | Realtime Data Llc | Bandwidth sensitive data compression and decompression |
US8090936B2 (en) | 2000-02-03 | 2012-01-03 | Realtime Data, Llc | Systems and methods for accelerated loading of operating systems and application programs |
US8275897B2 (en) | 1999-03-11 | 2012-09-25 | Realtime Data, Llc | System and methods for accelerated data storage and retrieval |
US8504710B2 (en) | 1999-03-11 | 2013-08-06 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US8692695B2 (en) | 2000-10-03 | 2014-04-08 | Realtime Data, Llc | Methods for encoding and decoding data |
US9143546B2 (en) | 2000-10-03 | 2015-09-22 | Realtime Data Llc | System and method for data feed acceleration and encryption |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0965171B1 (en) * | 1997-03-07 | 2005-11-30 | Intelligent Compression Technologies | Data coding network |
US6075470A (en) * | 1998-02-26 | 2000-06-13 | Research In Motion Limited | Block-wise adaptive statistical data compressor |
US5969647A (en) * | 1998-05-13 | 1999-10-19 | Dbtel Incorporated | Method of compressing letters in a caller identification code system |
US7609722B2 (en) * | 2003-02-14 | 2009-10-27 | Atheros Communications, Inc. | Method and apparatus for transmitting and receiving compressed frame of data over a wireless channel |
US9507811B2 (en) * | 2008-12-22 | 2016-11-29 | Oracle International Corporation | Compressed data page with uncompressed data fields |
US8188753B2 (en) | 2009-02-18 | 2012-05-29 | Analog Devices, Inc. | Analog computation |
EP3133737B1 (en) | 2010-08-18 | 2019-05-01 | Analog Devices, Inc. | Charge sharing analog computation circuitry and applications |
WO2012170533A2 (en) | 2011-06-06 | 2012-12-13 | Analog Devices, Inc. | Charge sharing time domain filter |
US9330148B2 (en) * | 2011-06-30 | 2016-05-03 | International Business Machines Corporation | Adapting data quality rules based upon user application requirements |
US9515679B1 (en) | 2015-05-14 | 2016-12-06 | International Business Machines Corporation | Adaptive data compression |
CN108600112A (en) * | 2018-05-08 | 2018-09-28 | 南京思达捷信息科技有限公司 | A kind of monitoring device and its method of cosmetics |
CN108650256A (en) * | 2018-05-09 | 2018-10-12 | 南京思达捷信息科技有限公司 | A kind of acquirement system of big data |
CN108916988A (en) * | 2018-05-09 | 2018-11-30 | 南京思达捷信息科技有限公司 | A kind of big data based on air obtains method |
CN110808738B (en) * | 2019-09-16 | 2023-10-20 | 平安科技(深圳)有限公司 | Data compression method, device, equipment and computer readable storage medium |
Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3394352A (en) * | 1965-07-22 | 1968-07-23 | Electronic Image Systems Corp | Method of and apparatus for code communication |
US3772654A (en) * | 1971-12-30 | 1973-11-13 | Ibm | Method and apparatus for data form modification |
US4355306A (en) * | 1981-01-30 | 1982-10-19 | International Business Machines Corporation | Dynamic stack data compression and decompression system |
US4612532A (en) * | 1984-06-19 | 1986-09-16 | Telebyte Corportion | Data compression apparatus and method |
US4642793A (en) * | 1983-03-24 | 1987-02-10 | International Computers Limited | Many-to-one mapping hash address generator |
US4701744A (en) * | 1986-03-27 | 1987-10-20 | Rca Corporation | Method and apparatus for compacting and de-compacting text characters |
US4804959A (en) * | 1987-11-10 | 1989-02-14 | International Business Machines Corporation | Method and apparatus using multiple codes to increase storage capacity |
US4843389A (en) * | 1986-12-04 | 1989-06-27 | International Business Machines Corp. | Text compression and expansion method and apparatus |
US5003307A (en) * | 1989-01-13 | 1991-03-26 | Stac, Inc. | Data compression apparatus with shift register search means |
US5016009A (en) * | 1989-01-13 | 1991-05-14 | Stac, Inc. | Data compression apparatus and method |
US5043728A (en) * | 1988-11-02 | 1991-08-27 | Canon Kabushiki Kaisha | Predictive coding device and predictive decoding device |
US5045852A (en) * | 1990-03-30 | 1991-09-03 | International Business Machines Corporation | Dynamic model selection during data compression |
US5229863A (en) * | 1990-12-24 | 1993-07-20 | Xerox Corporation | High speed CCITT decompressor |
US5369405A (en) * | 1992-05-19 | 1994-11-29 | Goldstar Co., Ltd. | Coefficient generation apparatus for variable length decoder |
US5386211A (en) * | 1993-09-29 | 1995-01-31 | Intel Corporation | Method and apparatus for recording and reconstructing binary data in a compressed form |
US5426779A (en) * | 1991-09-13 | 1995-06-20 | Salient Software, Inc. | Method and apparatus for locating longest prior target string matching current string in buffer |
US5493407A (en) * | 1990-11-29 | 1996-02-20 | Nec Corporation | Encoder in facsimile apparatus generates code words with one bit address indicating remaining bits are raw unencoded data when number of pixels in a run length are below a predetermined number |
-
1994
- 1994-12-06 US US08/350,426 patent/US5684478A/en not_active Expired - Fee Related
-
1997
- 1997-08-21 US US08/915,961 patent/US5949355A/en not_active Expired - Fee Related
Patent Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3394352A (en) * | 1965-07-22 | 1968-07-23 | Electronic Image Systems Corp | Method of and apparatus for code communication |
US3772654A (en) * | 1971-12-30 | 1973-11-13 | Ibm | Method and apparatus for data form modification |
US4355306A (en) * | 1981-01-30 | 1982-10-19 | International Business Machines Corporation | Dynamic stack data compression and decompression system |
US4642793A (en) * | 1983-03-24 | 1987-02-10 | International Computers Limited | Many-to-one mapping hash address generator |
US4612532A (en) * | 1984-06-19 | 1986-09-16 | Telebyte Corportion | Data compression apparatus and method |
US4701744A (en) * | 1986-03-27 | 1987-10-20 | Rca Corporation | Method and apparatus for compacting and de-compacting text characters |
US4843389A (en) * | 1986-12-04 | 1989-06-27 | International Business Machines Corp. | Text compression and expansion method and apparatus |
US4804959A (en) * | 1987-11-10 | 1989-02-14 | International Business Machines Corporation | Method and apparatus using multiple codes to increase storage capacity |
US5043728A (en) * | 1988-11-02 | 1991-08-27 | Canon Kabushiki Kaisha | Predictive coding device and predictive decoding device |
US5003307A (en) * | 1989-01-13 | 1991-03-26 | Stac, Inc. | Data compression apparatus with shift register search means |
US5016009A (en) * | 1989-01-13 | 1991-05-14 | Stac, Inc. | Data compression apparatus and method |
US5045852A (en) * | 1990-03-30 | 1991-09-03 | International Business Machines Corporation | Dynamic model selection during data compression |
US5493407A (en) * | 1990-11-29 | 1996-02-20 | Nec Corporation | Encoder in facsimile apparatus generates code words with one bit address indicating remaining bits are raw unencoded data when number of pixels in a run length are below a predetermined number |
US5229863A (en) * | 1990-12-24 | 1993-07-20 | Xerox Corporation | High speed CCITT decompressor |
US5426779A (en) * | 1991-09-13 | 1995-06-20 | Salient Software, Inc. | Method and apparatus for locating longest prior target string matching current string in buffer |
US5369405A (en) * | 1992-05-19 | 1994-11-29 | Goldstar Co., Ltd. | Coefficient generation apparatus for variable length decoder |
US5386211A (en) * | 1993-09-29 | 1995-01-31 | Intel Corporation | Method and apparatus for recording and reconstructing binary data in a compressed form |
Cited By (63)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8502707B2 (en) | 1998-12-11 | 2013-08-06 | Realtime Data, Llc | Data compression systems and methods |
US8643513B2 (en) | 1998-12-11 | 2014-02-04 | Realtime Data Llc | Data compression systems and methods |
US8717203B2 (en) | 1998-12-11 | 2014-05-06 | Realtime Data, Llc | Data compression systems and methods |
US8933825B2 (en) | 1998-12-11 | 2015-01-13 | Realtime Data Llc | Data compression systems and methods |
US9054728B2 (en) * | 1998-12-11 | 2015-06-09 | Realtime Data, Llc | Data compression systems and methods |
US20160261278A1 (en) * | 1998-12-11 | 2016-09-08 | Realtime Data, Llc | Data compression systems and method |
US7714747B2 (en) | 1998-12-11 | 2010-05-11 | Realtime Data Llc | Data compression systems and methods |
US20060181441A1 (en) * | 1998-12-11 | 2006-08-17 | Fallon James J | Content independent data compression method and system |
US10033405B2 (en) * | 1998-12-11 | 2018-07-24 | Realtime Data Llc | Data compression systems and method |
US7352300B2 (en) * | 1998-12-11 | 2008-04-01 | Realtime Data Llc | Data compression systems and methods |
US7161506B2 (en) * | 1998-12-11 | 2007-01-09 | Realtime Data Llc | Systems and methods for data compression such as content dependent data compression |
US20070109154A1 (en) * | 1998-12-11 | 2007-05-17 | Fallon James J | Data compression systems and methods |
US20030174732A1 (en) * | 1999-03-01 | 2003-09-18 | Hughes Electronics Corporation | Technique for data compression by decoding binary encoded data |
US7936787B2 (en) * | 1999-03-01 | 2011-05-03 | The Directv Group, Inc. | Technique for data compression by decoding binary encoded data |
US9116908B2 (en) | 1999-03-11 | 2015-08-25 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US10019458B2 (en) | 1999-03-11 | 2018-07-10 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US8719438B2 (en) | 1999-03-11 | 2014-05-06 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US8756332B2 (en) | 1999-03-11 | 2014-06-17 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US8504710B2 (en) | 1999-03-11 | 2013-08-06 | Realtime Data Llc | System and methods for accelerated data storage and retrieval |
US8275897B2 (en) | 1999-03-11 | 2012-09-25 | Realtime Data, Llc | System and methods for accelerated data storage and retrieval |
US6717580B1 (en) * | 2000-02-02 | 2004-04-06 | America Online, Inc. | Method and apparatus using a list of <minimum, size, gap, offset > quadruple to encode unicode characters in an upper/lowercase mapping |
US9792128B2 (en) | 2000-02-03 | 2017-10-17 | Realtime Data, Llc | System and method for electrical boot-device-reset signals |
US8112619B2 (en) | 2000-02-03 | 2012-02-07 | Realtime Data Llc | Systems and methods for accelerated loading of operating systems and application programs |
US8880862B2 (en) | 2000-02-03 | 2014-11-04 | Realtime Data, Llc | Systems and methods for accelerated loading of operating systems and application programs |
US8090936B2 (en) | 2000-02-03 | 2012-01-03 | Realtime Data, Llc | Systems and methods for accelerated loading of operating systems and application programs |
US20060190644A1 (en) * | 2000-02-03 | 2006-08-24 | Fallon James J | Data storewidth accelerator |
US6380873B1 (en) * | 2000-06-30 | 2002-04-30 | Quantum Corporation | Method for reducing radio frequency emissions on high-speed serial buses |
US20030164020A1 (en) * | 2000-07-29 | 2003-09-04 | Haberkamm Klaus Dieter | Method and device for band-edge orientated displacement of intermediate cylinders in a 6 cylinder frame |
US8742958B2 (en) | 2000-10-03 | 2014-06-03 | Realtime Data Llc | Methods for encoding and decoding data |
US10419021B2 (en) | 2000-10-03 | 2019-09-17 | Realtime Data, Llc | Systems and methods of data compression |
US9967368B2 (en) | 2000-10-03 | 2018-05-08 | Realtime Data Llc | Systems and methods for data block decompression |
US9143546B2 (en) | 2000-10-03 | 2015-09-22 | Realtime Data Llc | System and method for data feed acceleration and encryption |
US9859919B2 (en) | 2000-10-03 | 2018-01-02 | Realtime Data Llc | System and method for data compression |
US9667751B2 (en) | 2000-10-03 | 2017-05-30 | Realtime Data, Llc | Data feed acceleration |
US7777651B2 (en) | 2000-10-03 | 2010-08-17 | Realtime Data Llc | System and method for data feed acceleration and encryption |
US8692695B2 (en) | 2000-10-03 | 2014-04-08 | Realtime Data, Llc | Methods for encoding and decoding data |
US10284225B2 (en) | 2000-10-03 | 2019-05-07 | Realtime Data, Llc | Systems and methods for data compression |
US8717204B2 (en) | 2000-10-03 | 2014-05-06 | Realtime Data Llc | Methods for encoding and decoding data |
US9141992B2 (en) | 2000-10-03 | 2015-09-22 | Realtime Data Llc | Data feed acceleration |
US8723701B2 (en) | 2000-10-03 | 2014-05-13 | Realtime Data Llc | Methods for encoding and decoding data |
US8929442B2 (en) | 2001-02-13 | 2015-01-06 | Realtime Data, Llc | System and methods for video and audio data distribution |
US9762907B2 (en) | 2001-02-13 | 2017-09-12 | Realtime Adaptive Streaming, LLC | System and methods for video and audio data distribution |
US8867610B2 (en) | 2001-02-13 | 2014-10-21 | Realtime Data Llc | System and methods for video and audio data distribution |
US8073047B2 (en) | 2001-02-13 | 2011-12-06 | Realtime Data, Llc | Bandwidth sensitive data compression and decompression |
US8934535B2 (en) | 2001-02-13 | 2015-01-13 | Realtime Data Llc | Systems and methods for video and audio data storage and distribution |
US8054879B2 (en) | 2001-02-13 | 2011-11-08 | Realtime Data Llc | Bandwidth sensitive data compression and decompression |
US10212417B2 (en) | 2001-02-13 | 2019-02-19 | Realtime Adaptive Streaming Llc | Asymmetric data decompression systems |
US8553759B2 (en) | 2001-02-13 | 2013-10-08 | Realtime Data, Llc | Bandwidth sensitive data compression and decompression |
US9769477B2 (en) | 2001-02-13 | 2017-09-19 | Realtime Adaptive Streaming, LLC | Video data compression systems |
US20030110296A1 (en) * | 2001-12-07 | 2003-06-12 | Kirsch Steven T. | Method and system for reducing network latency in data communication |
US7370120B2 (en) | 2001-12-07 | 2008-05-06 | Propel Software Corporation | Method and system for reducing network latency in data communication |
US20080015844A1 (en) * | 2002-07-03 | 2008-01-17 | Vadim Fux | System And Method Of Creating And Using Compact Linguistic Data |
US20100211381A1 (en) * | 2002-07-03 | 2010-08-19 | Research In Motion Limited | System and Method of Creating and Using Compact Linguistic Data |
US7809553B2 (en) | 2002-07-03 | 2010-10-05 | Research In Motion Limited | System and method of creating and using compact linguistic data |
US7269548B2 (en) | 2002-07-03 | 2007-09-11 | Research In Motion Ltd | System and method of creating and using compact linguistic data |
US20040006455A1 (en) * | 2002-07-03 | 2004-01-08 | Vadim Fux | System and method of creating and using compact linguistic data |
US8001175B2 (en) | 2004-10-22 | 2011-08-16 | International Business Machines Corporation | Facilitating server response optimization |
US20060090033A1 (en) * | 2004-10-22 | 2006-04-27 | International Business Machines Corporation | Facilitating Server Response Optimization |
US20110208920A1 (en) * | 2004-10-22 | 2011-08-25 | International Business Machines Corporation | Facilitating server response optimization |
US8539022B2 (en) | 2004-10-22 | 2013-09-17 | International Business Machines Corporation | Facilitating server response optimization |
US7420992B1 (en) | 2005-03-17 | 2008-09-02 | Packeteer, Inc. | Adaptive network traffic compression mechanism including dynamic selection of compression algorithms |
US7148824B1 (en) * | 2005-08-05 | 2006-12-12 | Xerox Corporation | Automatic detection of character encoding format using statistical analysis of the text strings |
US7649909B1 (en) | 2006-06-30 | 2010-01-19 | Packeteer, Inc. | Adaptive tunnel transport protocol |
Also Published As
Publication number | Publication date |
---|---|
US5684478A (en) | 1997-11-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5949355A (en) | Method and apparatus for adaptive data compression | |
JP3553106B2 (en) | Text compression driver construction method and input text string compression method | |
US5572206A (en) | Data compression method and system | |
US5613145A (en) | Stored string data with element data units and pointer data units in distinct subranges of values | |
US5798718A (en) | Sliding window data compression method and apparatus | |
US4626829A (en) | Data compression using run length encoding and statistical encoding | |
US6522268B2 (en) | Systems and methods for multiple-file data compression | |
US4955066A (en) | Compressing and decompressing text files | |
US6542640B1 (en) | Data compressing apparatus, reconstructing apparatus, and its method | |
EP0083393B1 (en) | Method of compressing information and an apparatus for compressing english text | |
US5374916A (en) | Automatic electronic data type identification process | |
US5467087A (en) | High speed lossless data compression system | |
AU596713B2 (en) | Text compression and expansion method and apparatus | |
US5109433A (en) | Compressing and decompressing text files | |
US5737733A (en) | Method and system for searching compressed data | |
US6876774B2 (en) | Method and apparatus for compressing data string | |
US5254990A (en) | Method and apparatus for compression and decompression of data | |
US5717393A (en) | Apparatus for data compression and data decompression | |
US5226148A (en) | Method and apparatus for validating character strings | |
Hsu et al. | Automatic synthesis of compression techniques for heterogeneous files | |
US5815096A (en) | Method for compressing sequential data into compression symbols using double-indirect indexing into a dictionary data structure | |
US5394143A (en) | Run-length compression of index keys | |
Chung et al. | Level-compressed Huffman decoding | |
EP0871294B1 (en) | Method and apparatus for compression and decompression of data | |
Fenwick | Symbol Ranking Text Compression with Shannon Recodings. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: GARY, JAMES H., ILLINOIS Free format text: RESTATED PROMISSORY NOTE AND SECURITY AGREEMENT;ASSIGNOR:CENNOID TECHNOLOGIES, INC.;REEL/FRAME:012407/0864 Effective date: 20011113 Owner name: MILBANK CORPORATION, ILLINOIS Free format text: RESTATED PROMISSORY NOTE AND SECURITY AGREEMENT;ASSIGNOR:CENNOID TECHNOLOGIES, INC.;REEL/FRAME:012407/0872 Effective date: 20011113 Owner name: MILBANK CORPORATION MONEY PURCHASE PLAN AND TRUST, Free format text: RESTATED PROMISSORY NOTE END SECURITY AGREEMENT;ASSIGNOR:CENNOID TECHNOLOGIES, INC.;REEL/FRAME:012407/0880 Effective date: 20011113 Owner name: MER ROUGE PROPERTIES, LLC, ILLINOIS Free format text: RESTATED PROMISSORY NOTE AND SECURITY AGREEMENT;ASSIGNOR:CENNOID TECHNOLOGIES, INC.;REEL/FRAME:012407/0888 Effective date: 20011113 |
|
AS | Assignment |
Owner name: COMPRESSION FINANCING, LLC, ILLINOIS Free format text: SECURITY AGREEMENT - CONTRIBUTION AGREEMENTS, OPERATING AGREEMENT, AND RESTATED PROMISSORY;ASSIGNORS:MER ROUGE PROPERTIES, LLC;MILBANK CORPORATION MONEY PURCHASE PLAN & TRUST;MILBANK CORPORATION;AND OTHERS;REEL/FRAME:012875/0159 Effective date: 20020321 |
|
REMI | Maintenance fee reminder mailed | ||
FPAY | Fee payment |
Year of fee payment: 4 |
|
SULP | Surcharge for late payment | ||
AS | Assignment |
Owner name: COMPRESSION FINANCING, LLC, ILLINOIS Free format text: CERTIFICATE OF SALE;ASSIGNOR:CENNIOD TECHNOLOGIES, INC.;REEL/FRAME:014624/0674 Effective date: 20030730 |
|
REMI | Maintenance fee reminder mailed | ||
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20070907 |